Evaluation of a simulation model for prediction of herbicide movement and persistence in soil

Abstract

Summary The movement and persistence of residues of propyzamide, linuron, isoxaben and R‐40244 were measured in a sandy loam soil in field experiments prepared in spring and autumn. None of the herbicides moved to depths greater than 12 cm in the soil during the winter period, following application in autumn, and none moved more than 6 cm in the soil, following application in spring. The general order of persistence of total soil residues was isoxaben > linuron = R‐40244 > propyzamide. Appropriate constants to describe the moisture and temperature dependence of degradation were derived from laboratory incubation experiments and used with measurements of the strengths of adsorption of the different herbicides by the soil, in a computer model of herbicide movement. The model, in general, gave good predictions of total soil residues, but overestimated herbicide movement, particularly in winter. Measurements of herbicide desorption from the soil at intervals, during a laboratory incubation experiment, demonstrated an apparent increase in the strength of adsorption with time. When appropriate allowance was made for these changes in adsorption in the computer model, improved predictions of the vertical distribution of the herbicide residues were obtained.